Reheating Furnace Temperature Modeling And Control Based On T-S Model

Reheating furnace is one of the most important facilities in steelrolling, and consumes great amount of energy. And Reheating Furnace temperature is one of the most important parameters in the rotary kiln process control. So it is essential toimprove heating efficiency and reduce energy consumption for steelindustry. Modern rolling mills are developing to be continuous,large-scale, fast produce, high quality, which deliver a higher demandfor the research of reheating furnace modeling and optimal control.Reheating furnace is a typicalcomplicated industry system. Complex thermodynamic, chemical andphysical change makes it a system with multi-variable, time variety,non-linearity, strong coupling, pure time delay and a lot ofdisturbance factor. According to the walking beam reheating furnace in this thesis, T-S model is applied to the research of the reheating furnace temperature forecasting. And the model’s structure and parameters are identified apart. FCM algorithm is sensitive to its initial value and liable to be trapped in a local minimum. So in this thesis subtractive clustering is used to determine the initial clustering center and then FCM is used to improve the clustering’s convergence speed. At first subtractive clustering FCM is used to identify antecedent parameter and structure. Then, consequent parameter is identified by least square. This model is simulated by MATLAB using the data gathering on-site. Simulation results show that this method is effective.Reheating furnace has the feature of complicated circumstances and varity of parameters, higher quality calcination temperature controller is needed. And it is difficult to control the calcination temperature by traditional methods. So, in this thesis a hybrid control mothed is given, which is made of multi-model predictive control and fuzzy control through a fuzzy weight regulator, and the two controllers switch smoothly by using fuzzy rules. It does not only take full advantage of multi-model predictive controller’s high quality, but also avoid large computation and bad real-time character because of solving quadratic programming problems repeatly. What’s more, when predictive model of predictive controller does not match the plant, the fuzzy controller could work well as assistant.Simulation results show that this hybrid controller solved the mismatching of models and avoided large computation and bad real-time character. This compound controller can effectively control reheating furnace temperature.